A novel hybrid algorithm that employs BP neural network (BPNN) and particle swarm optimization (PSO) algorithm is proposed for the kinematic parameter identification of industrial robots with an enhanced convergence response. The error model of the industrial robot is established based on a modified Denavit-Hartenberg method and Jacobian matrix. Then, the kinematic parameter identification of the industrial robot is transformed to a nonlinear optimization in which the unknown kinematic parameters are taken as optimal variables. A hybrid algorithm based on a BPNN and the PSO is applied to search for the optimal variables which are used to compensate for the error of the kinematic parameters and improve the positioning accuracy of the industrial robot. Simulations and experiments based on a realistic industrial robot are all provided to validate the efficacy of the proposed hybrid identification algorithm. The results show that the proposed parameter-identification method based on the BPNN and PSO has fewer iterations and faster convergence speed than the standard PSO algorithm.

No abstract

To improve the positioning accuracy of industrial robots and avoid using the coordinates of the end effector, a novel kinematic calibration method based on the distance information is proposed. The kinematic model of an industrial robot is established. The relationship between the moving distance of the end effector and the kinematic parameters is analyzed. Based on the results of the analysis and the kinematic model of the robot, the error model with displacements as the reference is built, which is linearized for the convenience of the following identification. The singular value decomposition (SVD) is used to eliminate the redundant parameters of the error model. To solve the problem that traditional optimization algorithms are easily affected by data noise in high dimension identification, a novel extended Kalman filter (EKF) and regularized particle filter (RPF) hybrid identification method is presented. EKF is used in the preidentification of the linearized error model. With the preidentification results as the initial parameters, RPF is used to identify the kinematic parameters of the linearized error model. Simulations are carried out to validate the effectiveness of the proposed method, which shows that the method can identify the error of the parameters and after compensation the accuracy of the robot is improved.

Insufficient stiffness of industrial robots is a significant factor which affects its positioning accuracy. To improve the positioning accuracy, a novel positioning error compensation method based on the stiffness modelling is proposed in this paper. First, the positioning errors considering the end load and gravity of industrial robots due to stiffness are analyzed. Based on the results of analysis, it is found that the positioning errors can be described by two kinds of deformation errors at joints: the axial deformation error and the radial deformation error. Then, the axial deformation error is modelled by the differential relationship of kinematics equations. The model of radial deformation error is deduced through the recurrence method and rotation transformation between joints. Finally, these two models are transformed into a Cartesian coordinate system, and a positioning error compensation method based on these two models is presented. Simulations based on the finite element analysis are implemented to verify the positioning error compensation method. The results show that the suggested method can efficiently predict the positioning error according to the gravity and loads, so that the positioning accuracy of industrial robots can be improved with the proposed method.

To improve the positioning accuracy of collaborative robots, a novel modeling and calibration method for collaborative robots is proposed based on position information and modified local product of exponentials (LPoE). The kinematic error model is derived from the kinematic model through differential transformation. To solve the problem of the high redundancy and complexity of the error model that is difficult to identify, the errors of the kinematic parameters are all transferred to the initial position and posture matrix of the local coordinate system, which simplifies the error model and improves the speed and accuracy of the identification calculation. However, the simplified error model still requires posture data which are very difficult to acquire in practice. For this reason, the position error is separated from the kinematic model, and an error model based on position data is established. Then, a kinematic calibration method of the collaborative robot based on position data is proposed, which simplifies the measurement process and improves the efficiency of calibration. The effectiveness of the method is verified by simulations and experiments on a six degree-of-freedom collaborative robot.